How is AI changing collaboration and communication?

Seriously, AI is quietly reshaping the way we collaborate and communicate, and sometimes I'm quite surprised myself. I remember the old days of team collaboration when people had to get together for meetings or email back and forth, and it really wasn't efficient. What about now? Take a look at the group chat feature that ChatGPT just launched, which directly integrates multi-person collaborative conversations with AI assistants, a change that simply catches people off guard!

Interestingly, the change is not just at the communication level. Structured output features like the one recently introduced by Claude API make data processing surprisingly smooth. Developers can now ensure that the API response exactly matches the predefined JSON format, which saves so much debugging time in real-world applications. One development team I know said they reduced project delivery time by a third because of this feature, which is not a half a star efficiency improvement.

Moment of qualitative change in collaboration efficiency

In the past, when doing research for a project, just organizing the information can consume most of the day. Now Google NotebookLM's Deep Research feature can automatically browse hundreds of websites to generate reports, but also supports a variety of file formats. I have tried to use this function to deal with a market research project, the original need for three days of work, and now an afternoon can be done, but also automatically generate a list of reference materials, this experience is really very different.

And look at the REFRAG technology that Meta has introduced, which improves retrieval efficiency by more than 30 times. Imagine what that means in our daily work. What used to be a half-day wait to look up information is now an almost real-time response. This increase in speed makes teamwork a different pace.

Innovation in communication

What's even more interesting to me is that AI is even starting to understand image information. notebookLM can now directly read the contents of the picture, handwritten notes, charts and diagrams are converted into text analysis. Last week, my colleague casually photographed the minutes on the whiteboard, AI can accurately recognize and organize the main points, which in the past is simply unimaginable.

But then again, technology is evolving so fast that sometimes it's a bit hard to keep up. As Andrew Ng said, the real bottleneck now becomes how to get timely user feedback. It's true that no matter how good a tool is, if you don't know how to use it, or if you don't use it well, the effect will be compromised.

But overall, these changes brought about by AI have really made collaboration and communication more efficient and intelligent. I think we may not be able to live without these AI assistants in the future, and they are becoming indispensable partners in our work. It's just that in this process, how to balance technology and humanized communication may need us to continue to explore.